Show Less
Open access

Federalism and Secession


Edited By Jorge Cagiao y Conde and Alain-G. Gagnon

The controversial issue of secession has received little attention from experts of federalism. The best federal studies either evade it or dismiss it in a few lines. However, the issue of secession has been present throughout the history of federations. This book is one of the first to explore the complex relationship between federalism and secession.

The authors whose work is presented here recognize the potential of federalism as a way to organize relations between several different states, peoples, nations or territories under the same government. However, they are not naïve or idealist about the ability of the federal idea to succeed in the complex situations in which it is applied. In some cases success seems assured (the United States, Switzerland, Germany, etc.), and the merits of federalism can be showcased. But there are also failures (the former Yugoslavia, or more recently Brexit) and semi-failures that have generated turbulence in recent years in devolutive systems (Scotland in the United Kingdom, Catalonia in Spain) or federative systems (Québec in Canada).

This book provides a nuanced portrait of the issue of secession in federal contexts and lays the groundwork for questioning the still too fragile legacy of the great thinkers of federalism.

Show Summary Details
Open access

3 Are federalism and secession really incompatible?: JORGE CAGIAO Y CONDE

←116 | 117→


Are federalism and secession really


Jorge Cagiao y Conde


Secession has always been a difficult topic for the specialists of federalism to address and has even, over time, become almost taboo. And, of course, when someone decides to break a taboo and courageously speaks out, the words often appear hasty or clumsy, given the underlying fear of offending the audience. It is easy to understand why, at this point, the person decides to cut the presentation short, to the relief of both speaker and audience.

It is this kind of hasty, embarrassed (and often contradictory)1 approach to secession that is found in the literature on federalism, whether militant2 or ←117 | 118→academic.3 It leads to a conclusion that has come to dominate most specialized studies: that federalism and secession are incompatible.

The incompatibility is explained as being logical, with the result that any properly constituted and federative system should be designed so as to reject the secession of a federated unit. In a federation, whether constituted through the aggregation of formerly independent states (such as the United States) or the disaggregation or decentralization of a single existing state (such as Belgium), the federated units should be unable to leave the federation, and should not be able to rely on the principle of federalism to legitimize their secessionist project. In addition, the small number of procedures to regulate secession identified through comparative public law studies should be considered as exceptions to a rule which is, and must be, strict. The suggestion is that these exceptions can only be “non-federal” in nature.

I believe that this approach is problematical, especially since the dominant theory on federalism can be astonishingly flexible in the way it analyses federative systems. For example, the notion of sovereignty constitutes the classical dividing line between federative systems under domestic law, on the one hand, and under international law on the other (federal state/confederation).4 However this dividing line, observed by most specialists in the field of federalism, is increasingly considered to ←118 | 119→be irrelevant in properly informed studies of the federative objective in certain cases because – apparently – there is no sovereignty in a federative system.5 The contradiction accepted, on this point, by the dominant doctrine can be upheld only by accepting a great deal of flexibility. The same can be said of the fact that the doctrine agrees to classify, as federative systems, decentralized unitary states such as Spain.6 Many other examples could be given.

At this point in the examination, we simply need to remember that the dominant theory can be flexible when it wants to be. When this is not the case, flexibility is replaced by rigidity, as illustrated by the question that interests us here: secession. As we have just seen, sovereignty either exists (in the distinction between federal state and confederation, it is fundamental) or does not exist (to distinguish between a federation and a state, it cannot exist), depending on the argument applied. This shows a great deal of flexibility. But, to emphasize the point once again, flexibility tends to disappear when in a given federation (despite being described as non-sovereign) a demand to secede is made by a federated unit: the doctrine insists that it must be rejected, if needed by invoking the sovereignty (and the power that comes with it) of the federation (or its people).

←119 | 120→

Everyone will probably agree that it is hard to move forward over solid ground and to achieve any certainty about the phenomena studied (here the relationship between federalism and secession) when key concepts in the academic debate are ambivalent and plastic for political reasons. A scientific observer must, of course, note this ambivalence and plasticity, but without incorporating them into the approach used to address the issue. Although sovereignty, for example, is used as an argument by a state’s decision-making authorities and politicians, and more broadly by academics, to refuse to respond to secessionist demands and then, sometimes in the same breath, to state that it is important to move towards greater political integration in the European Union because sovereignty is a concept that no longer describes the actual functioning of states (which are no longer sovereign or independent, but non-sovereign and interdependent),7 this is not a valid reason for a researcher in the field of political science or law to validate the message conveyed in a discourse that, obviously, has no scientific or analytic goal or intention. Is it even possible for states to be sovereign within their borders and non-sovereign in their relations with other states?8 Is it not ←120 | 121→a little strange, after a demand for secession is received, to respond or explain that there is no point in aspiring to a sovereignty that no longer exists, even though the political will that blocks the secessionist process is probably the best demonstration that state sovereignty is thriving? Clearly, if we are looking for straight answers to our questions about secession and federalism, and as pointed out by Olivier Beaud,9 we need to move away from the state-centred – and, we are tempted to add, nationalist – ideological bias10 that is currently the dominant focus in the doctrine produced in this area.

←121 | 122→

In this study, my goal is not to produce yet one more opinion for or against the difficult question of secession, but rather, in an approach that is both simpler and more far-reaching, to gather the scant empirical evidence and certainties that can be assembled on this topic. It is this evidence that will help us answer the question of whether or not federalism and secession are incompatible under internal public law, as the dominant theory holds.

Given that my analysis is limited to the known empirical facts and evidence to which researchers have access, it is probably advisable to say (1) a few words about my methodology before developing my argument in three parts, as follows: (2) secession as viewed by federal theorists; (3) secession in federal positive law; (4) secession as seen from the standpoint of what could be called “legal logic” or “legal linguistics”,11 which comes down to almost the same thing.

1. General approach

The dominant approach in federal studies consists, in general and in summarized (and perhaps slightly deformed) form, in scrutinizing known federative law and federal practices to identify a regular feature (the rejection of secession) and in seeing this regularity as proof of the logical incompatibility of federalism and secession. This description of the federative experience is essential in order to build a (normative) theory of federalism in which secession has no place. As proposed as part of the theory of federalism, this approach matches an obvious legal naturalism: the end result (the rejection of the possibility of secession) ←122 | 123→is deduced from a fact (that in practice, the general rule is a rejection of secession). Besides the fallacious reasoning (we have known since Hume that just because something is does not mean that it ought to be), the dominant approach to the question is problematical in that it tells us with certainty what federative systems do most frequently, without telling us why they do it or whether what they do matches the principles of federalism. However, we know that the evolution of federalism (an evolution of which rejection of secession forms a part) was significantly impacted by other political projects of the modern era, including nation-building, state-building and democratization. We know today that these projects have changed the way in which the federal idea was initially understood, and created a gap between federal practices and the principles of federalism.12 The dominant approach, by ignoring the constant adaptation of federalism to two of the hegemonic political and legal theories of our modern political system (the state and the nation), states how federative systems generally deal with the problem of secession in nation-states (or federation-nations), but fails to explain, in any satisfactory way, why they do what they do. Apparently, the reasons lie hidden in the principle of federalism, but after taking a close look this seems doubtful.

In the following pages we will not follow this dominant approach. To ensure that our proposal is understood, we need to specify how we intend to focus on our subject. First, we must specify how the words “federalism” and “secession” are to be considered. The meaning of “secession” does not pose any specific problem (it immediately evokes a separation, a withdrawal, or independence, whether unilateral and against the will of the state – this is the most common definition – or negotiated and legally defined), but we intend to emphasize a more flexible and open-ended definition of federalism in order to answer the question placed at the head of this chapter. Last, we need to explain briefly why the legal approach used, which aims to produce a scientific result and say real things about its object, also comes to resemble the dominant theses of political science, by showing that there is no reason to abandon their ←123 | 124→flexibility and open-mindedness and occasional variations when looking at the question of secession.

First, the words “federalism” and “secession”. This article will examine the compatibility between the ideas of federalism and secession, and also the compatibility between a federal structure (a federal state or federation as an institution) and the right of secession guaranteed by that structure. The word “federalism” needs to be understood as both an idea and as the concrete or institutional expression of that idea (a federation).13 Although the relation between federalism and federation is complex and context-dependent, formal federations “without federalism” are not especially rare,14 and our investigation here is based on the congruence of federalism and federation. The question in the title of this paper could be phrased as follows: when federations follow the principles of federalism (another vast subject!),15 can they accept that their constituent units have a right of withdrawal or secession?

Our second observation concerns the meaning of the word “federalism”. What are we talking about here? The approach outlined in the preceding paragraph appears to call for an open and flexible meaning, as found in the specialized literature in the fields of political and legal science. Even if we stick to a rigid or restricted definition of the federal idea from a theoretical point of view (centring on two or three key concepts), we cannot ignore that even in the federal systems whose “federal” label has never been doubted (United States, Switzerland, Canada, Germany, Australia), the federal idea is, in practice, expressed through different approaches to various fundamental aspects: attribution or non-attribution of powers; constitution-pact or constitution-statute; presence or absence of a territorially-based second federal chamber; and so on. Although the systems are convergent in terms of principles (but not completely), they can easily diverge in the way they are implemented. This means that even if our investigations lead to the conclusion that federalism and secession are incompatible in theory, this may not apply in practice. Obviously, our goal here is also to find out if they are compatible at the theoretical level.

←124 | 125→

My last methodological remark concerns the legal approach to the problem, which clearly has had the greatest impact on the evolution of the debate. Like any other word or concept, federalism and secession can be examined from a legal standpoint. This occurs when a court is asked to rule on their compatibility, as the Supreme Court of Canada16 did after the second referendum on Québec independence in 1998 (the referendum was held in 1995). The Supreme Court provided an interpretation of the problem, since it will have legal effect. We know, since the work of Kelsen, that this interpretation is not intended to extend knowledge or provide an objective explanation of the problem posed.17 A court rules on the legal problem brought before it, an intellectual exercise that requires it to assign, to each legal word or statement (“federalism”, for example) one out of a range of possible meanings in a given context. In other words, the court explains which of a range of possible versions it intends to select as the only legal version, the one that will have effect in law. We also know, once again thanks to Kelsen, that this way of settling a problem in law does nothing to settle the problem in terms of knowledge (or principles): other meanings, and therefore other responses or solutions to the same problem, remain possible.18

Although nothing in the above explanation will meet with much resistance from experts in the science of the law (teachers, researchers), at least in theory, it is important to note that when they actually examine a matter they often tend to act like judges when explaining a problem.19 ←125 | 126→If the question, for example, is whether federalism, properly understood, and secession, properly understood, are compatible (in terms of knowledge rather than in terms of decisions, since as we will see it is perfectly possible to decide not to make them compatible), they will argue that there is only one possible answer,20 rather than two or three. Their discourse is neither explanatory nor descriptive (analytical or scientific), but normative. Judges, of course, rule on a legal dispute because that is their job. They must state the law. Their discourse is the discourse of the law. However, this is not the case for scientists of the law: even if they want it to, their discourse is not the law, at least not in the sense that it makes the law. All they can do – and this is their function as teachers and researchers – is to explain and describe the operation of the law. Their discourse is not a discourse of the law, but a discourse about the law. They do not produce law (at least not directly, but can do so indirectly if they influence the law-makers), but knowledge of the law. If knowledge of the law is generally plural, as indicated by the scientific interpretation of Kelsen, we have every reason to believe that the answer to the question posed in this chapter will be an expression of the flexibility and plasticity that the political and legal scientific literature discovers when it turns its attention to these issues.

I can now turn to the question of what can be said about the federalism-secession relationship once positioned and examined using the tools described above.

2. Secession as seen by the theoreticians of federalism

This section will focus on the viewpoint of theoreticians of federalism concerning compatibility between federalism and secession. As mentioned above, the majority opinion sees an absolute incompatibility between the two terms. This view became dominant in the 19th century and was confirmed in the United States, after the Civil War (1861– 1865), by Texas vs Whyte21 in which the Supreme Court described the federation as an indestructible Union, composed of indestructible States. ←126 | 127→Doctrine then took up the task of validating this solution and, as pointed out by Laurent Déchâtre in a recent study, authors such as “Jellinek, Laband and Le Fur considered that the members of a federal state are in a subordinate relationship with the federal authorities, making their sovereignty impossible, and from this they deduce the exclusion of any right of secession.”22 This dominant thesis has strengthened over time and is now, today, in very good health. It appears to be followed even by authors who, like Olivier Beaud, offer far-reaching criticism of the dominant theses in studies of federalism. It is important to remember that Beaud’s theoretical construction views federalism on the basis of its contractualist or pactist foundation, with a double objective (union and particularism) that must guarantee states that commit voluntarily to a federal relation the freedom they initially enjoyed. From this point of view, Beaud’s intention is to distance the theory of the federation from the single-state version of federalism (federal state) that dominates federal studies and positive law based on the notion of “sovereignty”. According to Beaud, the notion of federation must be addressed without relying on sovereignty, which prevents us from understanding it properly. Beaud’s approach hits a major snag, however, when the federation is considered as a perpetual union, a sort of trap into which federated states fall voluntarily, since they lose “their sovereignty (in the strict sense) when they become member states”.23 At this point it is reasonable to ask, as mentioned above, if this conclusion does not let sovereignty in by the window after the door has been closed. If secession is no longer an option for the member states of a federation, then it can only mean that the federation disposes of all the resources that the federative system makes available (including the use of force) to prevent secession by a state. If this is true, then the federation is sovereign.24

However, it is not the dominant opinion that interests us here. We know it is does not support secession, and that support for this thesis and its various versions is widespread. We are more interested in the opposite ←127 | 128→position: does it have any defenders among theoreticians of federalism? This is an important question, since if it transpires that no theoretician or observer of federalism has considered the possibility of secession from a federation, we could close our investigation. In a hurry, we could conclude that the two terms are incompatible. But this is not the case, as we will show briefly here.

In reality, the idea that a federation is a free union of free and independent states, based on a compact, and that the sovereignty of the member states remains intact during the union, was relatively widespread in the United States prior to the Civil War.25 If it was taken for granted by the defenders of States’ Rights in the South like John Calhoun, it also had support among the supporters of the Union: in a free union, a state could leave the federation if it so wished. For example, this is what Thomas Jefferson stated in 1816: “If any State in the Union will declare that it prefers separation […] to a continuance in union […] I have no hesitation in saying, ‘let us separate’.”26 Of course, the Civil War provided an opportunity for everyone to demonstrate, by armed force, that secession was an act of rebellion that would be blocked by the federation. However, the theory in favour of the right of withdrawal was not necessarily in contradiction with the spirit of federalism, as recognized a few years later by Carl Schmitt, an observer unlikely to have secessionist sympathies. In his view, it involved “essential concepts of constitutional theory for a federation”.27 Secession had major support among the theoreticians of compact-based federalism in the United States.

At around the same time, another theoretician of federalism was defending the same thesis in France: Pierre-Joseph Proudhon, author of the well-known Du principe fédératif (1863). His contribution is especially interesting since it explicitly refers to the principles of federalism (Proudhon was writing in a country fiercely opposed to federalism, and his thoughts did not reflect a federal system and practices like those in ←128 | 129→the United States and Switzerland)28 and attempted to conceptualize federal rules and procedures (the federation) on the basis of those principles. Proudhon considered that the federation did not follow the logic of the state:29 “It is a group of sovereign and independent states.”30 These sovereign and independent states were linked by a federative pact, but according to Proudhon the question of perpetuity was not an issue. The pact could be created with that intention without depriving the states of their right to withdraw if they considered that the federation no longer protected their legitimate interests effectively. This was how he viewed the secessionist conflict in the United States and the Sonderbund War in Switzerland, in 1846, in both cases defending the legitimacy of the secessionist parties: “Separation operates by reason of law.”31 If the federation is not a state, if the federated states do not transfer their initial sovereignty to the federation, and if the federal relationship is not intended to modify the initial situation against the will of the federated units, then Proudhon’s conclusion appears to respect the initial logic: states federate, but not with the aim of losing their sovereignty.32 If they are still sovereign once the federation has been created, they should be able to leave in the same way they entered, by expressing their will to withdraw.

←129 | 130→

Here we have two theoretical contributions from federalists who considered federation and secession to be compatible. What remains of this position today? Has it been able to resist a century and a half of evolving federalism that has brought it closer to the state model, or must we conclude that it belongs to a 19th-century view of federalism that is completely outmoded today? In fact, the topical nature of secession in politics and comparative law has become ever more evident in recent decades, and it has also been enriched by new approaches and theoretical contributions following the collapse of the former USSR and the independent countries that have emerged. An essay by Allen Buchanan, Secession. The Morality of Political Divorce from Fort Sumter to Lithuania and Quebec, published in 1991,33 is clearly the most important of a growing list of works devoted to secession that do not lead to the immediate rejection and moral condemnation of the idea.34 Two main theories of secession have emerged: secession as reparation for a wrong (remedial secession) and secession as a primary right enjoyed by all peoples, and even any human group that requests it (without necessarily being a “people” or “nation”). Although the first theory considers secession as being morally founded only in certain cases (the difficulty being how to decide if there is a just cause for secession),35 the second appears to come closer to the position of the contractualist theoreticians of the 19th century who defended the moral nature of any secession desired by a people or nation (the difficulty being how to know which ←130 | 131→human groups or collectives should be considered as a people or nation for the purposes of secession).

One interesting point deserves to be highlighted in the academic literature that has developed on this question: the fact that secession can be considered a moral right36 or a morally legitimate aspiration. To simplify, this favourable (or at least not unfavourable) opinion of secession was formerly specific (for obvious partisan reasons) to independentist movements or leaders, but only a few non-independentist thinkers (Proudhon being one of them) defended the morality and legitimacy of secession. The situation has changed somewhat and the scientific literature today presents secession not only as a moral right held by certain peoples (or by all peoples, depending on whether the theoretician supports “remedial” secession or secession as a primary right), but also as a way to ensure the stability of the system and of the federal principles37 (loyalty, trust, fidelity, freedom, etc.) within a federation. The work of authors such as Daniel Weinstock,38 Wayne Norman39 or Miodrag Jovanović40 appears to point in this direction. Not only is the constitutionalization of secession not incompatible with federalism but, by not making this legitimate aspiration of a federated union to recover its full freedom non-viable, the federation can be made stronger: unity is not imposed, but desired. In addition, as the empirical evidence shows in democratic contexts such as Canada or the United Kingdom – Weinstock’s contribution also suggests this – the existence of a “constitutionalized” right of secession could create a danger for secessionist movements, since they could be defeated not by a federation logically opposed to their claims, but by their own people.41 The experience of Québec in 1995 shows that a referendum defeat, even extremely close, is not without consequences for an independentist movement. The two-edged sword of ←131 | 132→a secession clause would probably incite independentist movements to be more careful, aware that independentist processes in liberal democracies tend to involve an uphill battle, unlike non-democratic or violent contexts (Sorens describes these as a “secessionism of despair”).42 Seen from this standpoint, the constitutionalization of the right of withdrawal, rather than being an ideal instrument allowing independentists to secede easily, can be viewed as a dissuasive element and therefore as a factor that strengthens, rather than weakens, a federation.

Between the theoreticians of federalism in the 19th century and the present day, the defence of the right of withdrawal has taken on a new dimension. Not only are federalism and secession seen as compatible, but the former may need the latter to gain strength and achieve its aims. In any case, this brief survey of the theory of federalism shows that some theoreticians have been able to consider the two terms as being compatible, and even necessarily compatible. From this point of view, it is making them incompatible that appears to be incompatible with the principles of federalism, or at the very least counter-productive, since the federated units would then tend to feel their links with the federation at a particular time as being imposed rather than desired. Federalism, at least in theory, relies on the freedom of all the parties in the federative relationship.

3. Secession in positive law

The numerous constitutions of the past and present that either reject the possibility of secession or remain silent on the question have, inevitably, led to the development of jurisprudence and doctrine that deny the existence of a right of secession.43 In this section, we will attempt to discover if there are any federative systems that accept secession as a legal possibility. In addition, if it is possible to identify non-federative systems that accept the secession of part of their territory, this would probably support the compatibility of secession and federation since a federation, unlike a unitary state (centralized or decentralized), is already made up of distinct territorial units that may have been independent at some point in the past. In addition, it is generally agreed that unitary states are more ←132 | 133→rigid than federations with respect to territorial unity, since federations are both better prepared for the scenario of independence (having a complete and operational decentralized administration) and, in some cases, are built on the idea of consent (from the territorial units) that is generally absent in a unitary system. When the question of secession arises, it is indirectly the question of consent to being governed by the existing authorities that is being challenged.

We can set aside the historical examples (whose federal label also happens to be questionable) such as the former USSR, which recognized the right of secession of its territorial units while denying, de facto, the exercise of that right. We simply need to note the cases of secession that occurred unilaterally, and therefore illicitly under the reference legal framework, in the Eastern Bloc (in the former Soviet Union and Yugoslavia). With respect to compatibility between federalism and secession, it would be sufficient to observe one or more federative systems that have made the two concepts compatible to conclude on their compatibility. It could be considered that the two cases just mentioned are enough to demonstrate compatibility. However, we can try to delve deeper (doubts about the democratic label encourage this endeavour) and a little closer to home, by briefly reviewing some contemporary cases.

Saint Kitts and Nevis (two islands in the Caribbean) and Ethiopia are, to date, the only federations to have included a procedure for secession in their constitutions, and their federated units can therefore secede legally. Article 113 of the Constitution of Saint Kitts and Nevis provides for Nevis to separate from Saint Kitts: “The Nevis Island Legislature may provide that the island of Nevis shall cease to be federated with the island of Saint Christopher and accordingly that this Constitution shall no longer have effect in the island of Nevis.” It is important to note that the procedure set out in the following articles makes secession especially difficult (it has already been attempted once without success). In Ethiopia, the constitution states as follows (article 39.1): “Every Nation, Nationality and People in Ethiopia has an unconditional right to self-determination, including the right to secession.” In reality, repression of the Oromos (who make up between 20 % and 30 % of the federation’s population) by the Ethiopian government in recent years creates some doubt about the conditional (or even unconditional) nature of the right of secession, but this cannot erase the existence of a constitutional clause providing for a right of unilateral secession.

←133 | 134→

As we have just seen, recognition of a right of secession is apparently not incompatible with the pursuit of the federation’s initial objectives. However, we should note that these two cases are often presented in political and academic debates as examples that are not particularly relevant, since politicians and researchers from liberal democracies like to point out that both countries have a history and democratic label of questionable quality. We note the objection, but nevertheless observe that two systems with a questionable democratic label have been able to make federalism and secession compatible by including both terms in their constitution. The objection is noted, and since our main interest is the compatibility between federalism and secession in democratic systems, we must attempt to find something better elsewhere.

Looking at the Western democracies, the case that has received the most attention (and is probably the most interesting) in recent years is the case of Canada. The question of secession arose in the province of Québec for the first time in 1980, and for the second time in 1995. Both referendums on Québec’s independence were won by the “No” side, the second time on a slim majority.44 The Canadian constitution is silent on the question of secession. Developments in the areas of doctrine and jurisprudence tend to consider Canada as an indivisible political unit, like its neighbour to the south (“indestructible”). Given that Québec (like the other provinces) can ask its citizens to give their opinion on the question of secession in a legal referendum (this is a provincial power), the silence of the constitution, even when interpreted as indicated, could be a problem if the “Yes” side won a third referendum (which has not yet taken place). After the extremely close referendum result of 1995, the Canadian government became involved and asked the Supreme Court of Canada to rule on the constitutionality of a possible Québec secession.

The Supreme Court was asked to examine the question from the standpoint of Canadian constitutional law and international public law. The questions were worded as follows:

“1. Under the Constitution of Canada, can the National Assembly, legislature or government of Quebec effect the secession of Quebec from Canada unilaterally?

 2. Does international law give the National Assembly, legislature or government of Quebec the right to effect the secession of Quebec ←134 | 135→from Canada unilaterally? In this regard, is there a right to self-determination under international law that would give the National Assembly, legislature or government of Quebec the right to effect the secession of Quebec from Canada unilaterally?

 3. In the event of a conflict between domestic and international law on the right of the National Assembly, legislature or government of Quebec to effect the secession of Quebec from Canada unilaterally, which would take precedence in Canada?”45

The Supreme Court concluded that a unilateral secession was unconstitutional and that there was no legal framework in international law that would make it legal. It noted, however, that its review of Canadian constitutional law clearly revealed the existence of four principles (federalism, democracy, constitutionalism and the rule of law, respect for minorities) that underlie the Canadian constitutional order and which, correctly interpreted and balanced, require the Canadian government not to ignore a clear referendum vote in favour of secession and therefore to negotiate a solution in good faith with the Québec government: “The continued existence and operation of the Canadian constitutional order cannot remain indifferent to the clear expression of a clear majority of Quebecers that they no longer wish to remain in Canada.”46

The solution indicated by the Supreme Court would involve negotiations to amend the constitution to respond to the legitimate claims of the secessionist territory.47 If the two parties disagreed and the negotiations failed, whether in good or bad faith, secession would fall outside the legal framework, which does not necessarily mean, in the view of the Supreme Court, that it would have no democratic legitimacy.48 The Supreme Court’s Advisory Opinion was followed immediately by an Act of the Canadian parliament, the so-called Clarity Act, in 2000,49 which can be seen as constitutionalizing the right of secession by providing for a procedure which, although not entirely in the spirit of the Supreme ←135 | 136→Court’s Advisory Opinion,50 lays the foundations for a right of withdrawal from the Canadian federation. The Québec parliament responded to the Canadian Act with its own Act in the same year (2000),51 in which it expresses its disagreement with the federal government’s interpretation of the Supreme Court’s Advisory Opinion of 1998.52

Whatever happens in connection with Québec secession in the future, the Canadian federation now has a procedure and tools that confirm the possibility of secession in a federative system. In Canada, one of the world’s most dynamic federations that is also one of the most closely studied, federalism and secession do not appear to be considered to be incompatible.

Recognition of secession is not just a stop-gap solution or an anomaly found in poorly-integrated democratic contexts, which are more exposed to the risk of separatism, like Brexit in the EU53 or – as some might say – the case of Québec in Canada. The proof of this is that it is even possible to find unitary states (whether centralized or decentralized) that have agreed to recognize the right of secession. Examples include the centralized states of Uzbekistan, Denmark and France, which have constitutional provisions in this area. Uzbekistan has recognized the right of secession of the Republic of Karakalpakstan (article 74): “The ←136 | 137→Republic of Karakalpakstan shall have the right to secede from the Republic of Uzbekistan on the basis of a nation-wide referendum held by the people of Karakalpakstan.”54 Denmark, in turn, supported the process that allowed Greenland to gain greater autonomy, which culminated in a victorious referendum in 2008 (a 75 % vote in favour in Greenland) and the Act on Greenland Self-Government (2009)55 that provided for the possibility of accession to independence in the following terms (article 21): “Decision regarding Greenland’s independence shall be taken by the people of Greenland.” France, which experienced secessionist events in the past (Algeria, Comoros),56 has enshrined a right of secession (self-determination) for New Caledonia in the constitution of the Fifth Republic (articles 76 and 77). The objection could be made that the right to self-determination exists in international law for colonies, and that France has merely recognized and integrated this international right in its constitution to allow colonized peoples (as in the cases mentioned previously) to achieve self-determination. This is accurate, but we fail to see what difference it makes, given that our focus is on the compatibility between the unity of a political body (article 1) and the secession of one of its parts, provided for New Caledonia in articles 76 and 77 and for all other territories in article 53.57 Neither the specific historical context (colonization and then decolonization) nor the legal source (in this case, international law transposed into domestic law) have a significant impact on our conclusion. They simply explain the reasons for which a system agrees to make unity and secession compatible: an occupation felt to be unjust, a higher international norm that a country agrees (with emphasis on “agrees”) to follow, and so on. The conclusion remains the same: the indivisible unity enshrined in the constitution does not appear to be incompatible with the possibility of secession in constitutional law.

←137 | 138→

To conclude this section, with regard to decentralized states, we have the example of the referendum on independence for Scotland in 2014. The British constitutional system is not thought of as being less united or consolidated than others, but the British government was able to find a negotiated outcome with the government of Scotland when, faced with the same problem, the government of Spain (another decentralized state) did not consider it advisable to grant the request of the Catalan secessionist movement, which wanted to organize a referendum on independence based on the model of Scotland or Québec.58 Once again we have two examples that show that the law can make unity and secession either compatible or incompatible, depending on the wishes of the authorities or stakeholders making the decision.

All of the above tends to confirm that there is not necessarily any incompatibility between federalism and secession in positive law. Constitutional law can, of course, exclude a territory’s right of withdrawal, but nothing seems to stop a federation or state from recognizing such a right, as the examples in this section show.

4. Secession and “legal logic”

As mentioned previously, a recurrent argument in the debate about secession highlights its incompatibility with the unity of the state or nation, if unity is explicitly mentioned in the constitution.59 Examples include article 1 of the constitution of the Fifth Republic in France (which is “indivisible”) and article 2 of the Spanish constitution (“the Constitution is based on the indissoluble unity of the Spanish nation”). A similar observation could be made about any state or federation that ←138 | 139→is considered to have been established in perpetuity. This applies, as outlined above, to way in which the dominant doctrine in the United States interpreted the constitutional pact of 1787 almost unanimously after the Civil War. In law, apparently, this type of constitutional provision has the effect of making it almost impossible for the legal system to deal with secession.

However, we should note that this reasoning appears to abandon the method of legal positivism in favour of another form of positivism characteristic of legal systems,60 which tend to present the legal order as a perfectly coherent entity, based on a certain number of concepts or principles that, provided one uses the spectacles offered by the system, reveal the completeness of the legal order, the normative hierarchy, legal security, the separation of powers, etc.61 According to the positivist discourse (approach) of legal systems (the second meaning proposed by Bobbio: “a theory”),62 legal orders are “coherent systems”.63 Based on this specific logic, the contradiction between secession and unity is clear and, of course, must be denounced.

There are two things that need to be highlighted about this approach and (self-proclaimed) description of the law. First, it is important to note that by adopting the discourse and logic of the legal system, researchers intervening in the debate also adopt the point of view and language of the legal order (its authorities and players). And this, as we have mentioned, is the language of “ought”, of what ought or ought not to be.64 When the law prohibits an action or omission, and prescribes the related penalty, ←139 | 140→it is not describing an actual fact, but ordering a course of conduct (to do, or not to do, as the case may be; to punish a person who does what is prohibited, or fails to do what is prescribed). Next, if the legal system wants certain rights or values to have special protection or be strengthened (for example, unity of the state or fundamental rights), it may find it useful to consider that this duty will be better protected from non-compliance if citizens no longer view the norms through which the legal system expresses and enacts its will as an intention (what “ought” to be), but rather as what already “is”. That which can only be expressed legally as an “ought” is presented by the system as if it were an observable empirical fact. The unity of the state, for example, mentioned in some constitutions, is seen, in this special legal view, not as an intention or something that “ought” to be, but as the observation or reporting of a natural fact, such as falling rain or the sun setting in the west. In this approach, the law simply “observes” a fact (seen as being independent of law), allowing a shift towards a form of reasoning in which the players make statements with the same assurance as if talking about the empirical world. For example, a statement (legal, in this instance), can be described as true or false in the same way as the statement “Peter is in the room” is false (if he is not there) or true (if he is). In this representation, the unity of the state or federation is “ontological”, something that “is”. When the French republic is described as being indivisible, the statement does not express an intention or duty to protect the unity of the political body, or an objective assigned to the instituted authorities by the constituent power, but describes an observable empirical fact: the political body is singular and indivisible. And legal specialists discussing the norm, as legal specialists, can say: it is true, the republic is indivisible, with the same assurance as if they were saying that Peter is in the room (if he is).

It is not hard to understand that this thesis affirms something that is false and almost absurd: the republic may be united when the constitution is created, and there may be a wish for it to be protected from attempts to disunite it or cause it to collapse, without it being possible to reach a conclusion as to its effective or real indivisibility. In fact, the opposite is true: the republic can be divided, into as many parts as can be imagined. This is shown by the examples above. The legal discourse, a discourse of what “ought” to be, has no power to change anything in the world of what “is”: something that is naturally divisible remains divisible whatever the law says (a territory – a space, and the power that is deployed over that space – is always divisible).

←140 | 141→

Based on this explanation, we can better understand the need for researchers to abandon the approach of legal positivism, specific to legal systems, and to adopt positivism as the approach used to focus on its object of study (Bobbio). This attitude or gaze, when directed at the law, shows that the legal discourse is actually a normative discourse that expresses a will, or what “ought” to be, and is not a description or explanation of reality. Using this first observation (the law is expressed in the form of “ought”), we can ask if a legal order can accept two intentions or “oughts” that are contradictory in appearance, such as indivisible unity and secession. This is a question that, at first glance, we are probably inclined to answer with a “No”. A and non-A cannot exist at the same time, we might think. The situation is, however, a little more complex. First, because although it is true that one cannot be at the oven and at the mill at the same time (an empirical evidence), it is also true that a legal system (or natural person) can want two things at the same time: for example, to protect its territorial unity and to recognize the right of secession of its constituent territories. It is even possible, as we saw above, to find a coherent explanation for this combination: we want unity to be, but also to be wanted by the various territorial entities; those that do not want unity can then be asked to constitute themselves separately. Secondly, because the two intentions or rights (because the same thing can be expressed in terms of rights) are not held by the same parties: the state (or the people) holds the right to territorial integrity, while a part of the state’s territory holds the right to constitute a separate state. In fact, the same applies to other rights: our right to express our opinions freely must not conflict with the right of others to see their honour protected, for example. The rights are different, and so are the holders. But they may still conflict. For unity and secession, the situation is the same: the right of the state to protect its unity or territorial integrity can be accommodated with a recognized right for the populations inhabiting the constituent territories to leave. We have identified examples in the field of comparative public law that confirm this is possible.

Last, we can follow another path, which we could call the Kelsenian path, to see the incompatibility we are investigating as a conflict of norms65 rather than rights. The norm of “indivisible unity” and the ←141 | 142→norm of “secession” conflict with each other; and Kelsen states that “A conflict exists between two norms when that which one of them decrees to be obligatory is incompatible with that which the other decrees to be obligatory”.66 The conflict must be viewed dynamically (as a conflict of norms, when the exercise of the right of withdrawal is opposed by the norm that affirms indivisible unity) and not statically or in the abstract (based on the two “contradictory” norms in the constitution) since, in the latter case, there is no actual conflict.67 However, we can still explore this avenue quickly. As Kelsen explains, the conflict of norms occurs only between valid legal norms, and the resolution of the conflict by the competent authority does not suppress the validity of the norm that is set aside: “the situation created by a conflict of norms is that one of the two conflicting norms is being observed and the other violated, not that only one of the two may be valid.”68 The two norms are therefore both still valid, including the norm that cannot be observed in the specific case that led to the conflict. This means, for example, that if a territory secedes from a state or federation, the norm that affirms the indivisible unity of the state or federation is still valid and still has effect within its territory. If, on the other hand, secession is prevented by the state or federation on the basis of the “unity” norm, the “secession” norm remains valid and may be applied successfully at a later date. This leads to the conclusion that the legal contradiction or incompatibility exists in appearance only, if the system provides for mechanisms to ensure that two apparently contradictory norms cannot both apply at the same time; instead, one may apply in some cases while the other applies in other cases.

In fact, even if this specific relation (the simultaneous presence of unity, presented as indestructible, and of a recognized right of secession) is seen as a clear contradiction, we still need to recognize that legal systems contain this type of contradiction and have the mechanisms needed to ensure that something that appears incompatible or contradictory becomes legally compatible and coherent within the system and in the eyes of the public. There are two possibilities: either there is no contradiction ←142 | 143→between unity and secession in the cases presented above, or there is a contradiction, but the systems accommodate the contradiction by denying that it occurs, which is a double contradiction. In both cases, cohabitation between unity and secession is possible.

Although incompatibility must be identified between the obligations, rights and norms (or rather legal statements) that appear to contradict each other, legal logic or linguistics – the way in which the law expresses itself – appears to indicate that no incompatibility exists. If, as pointed out by Kelsen, the sign of a logical contradiction is that it exists between statements or judgements that are true or false, but not between norms, because norms are not true or false but only valid or invalid. However, the contradiction in our case arises between two norms (legal statements) or rights (a right takes the form of a legal statement) that are valid, in other words belong to the same legal system. From this point of view, there simply cannot be a contradiction or incompatibility.

We still need to examine another objection that is sometimes raised against the idea that federalism (unity) and secession can be compatible. This time, it is not the right of secession that is targeted and defined as being incompatible with the right to unity of the federal political body, but actual secession (the fact of secession). According to various authors, secession destroys federal (or state) unity, and therefore destroys the federation itself. Under this argument, when actual secession is illegal (this argument has re-emerged lately in the unionist camp in Spain), it also destroys democracy and the state of law. We will focus on the first of these objections, clearly the most serious: that secession destroys the federation (or state). First, we must examine a technical argument that is nevertheless important: if unity is expressed as an “ought”, as we saw above, then it cannot be harmed by an “is”, an empirical fact such as secession. We can consider the following example: a thief enters a jewellery store and leaves with an impressive haul. Neither the police nor the justice system reacts. Is it the theft that violates the penal norm, or the fact that the offence is not punished?69 We could say both, because an offence is needed (the condition) for there to be a penalty, but this answer is not completely satisfactory. Or, we could take another approach and ask if the violated norm is still valid, if it continues to play its role in the legal ←143 | 144→system. The answer, if it has not been repealed (and it is clearly not the theft that can repeal the penal norm), is clearly “Yes”. The same answer applies for the norm that states the requirement of state unity. Unity is required, but a territory has decided to leave the federation. Will the norm still be valid for the remaining territories in the federation? Nothing allows us to believe that it will not apply. The only case, we believe, where it is possible to consider that actual secession may be incompatible with federalism, or rather with the existence of the federation or state, is the case in which the federation loses most of its member territories (such as the former Yugoslavia) or has only two territories (such as Belgium). In the latter case, the departure of one of the two partners in the federation or state would automatically lead to the end of the federal relationship. Actual secession would apparently end the federation, which could be seen as a factual (but not legal) incompatibility: the existence of secession means the inexistence of the federation, and vice versa. The question of whether, in the specific case of a two-party federation that ceases to exist because one of the parties secedes, federalism (the idea) and secession are also incompatible may, however, lead to a different answer.


The goal of this article was to examine the question of whether federalism and secession are compatible, by highlighting the experiences and facts (including linguistic facts) that can be used to answer the question objectively.

Our investigation leads to the conviction that nothing supports the conclusion that the two terms are logically incompatible. Clearly, a federation (or state) can refuse to enshrine a right of secession in its constitution for its member territories, or may decide not to grant that right if the constitution remains silent on the question. This attitude is not necessarily incompatible with federalism (or at least a certain kind of federalism) or the existence of a federation, as shown by the numerous authors who have viewed federalism as a perpetual union, as well as the numerous legal systems built and developed on the same foundation and dynamic. However, the opposite solution, for the same reasons, leads to the same conclusion. The federal idea has been supported by theoreticians respected and esteemed for their strong and original contribution to federal thought, while at the same time defending the fundamental need to include a right of withdrawal. And although we ←144 | 145→have seen that it is common for federative systems under public law to refuse of a right of secession, we have also observed that secession remains a legally-considered solution in some federations (Ethiopia, Canada) and even some unitary or decentralized states (Denmark, France, UK). It is not even clear that federations can be distinguished from unitary states on this point, since everything supports the conclusion that secession is possible in both types of state.

Last, “legal logic”, using the approach we have applied in this chapter, shows that the contradiction or incompatibility that the doctrine identifies between federalism (the unity of the federation) and secession is far from clear. Of course, from the point of view of the system (what we have called the “discourse of the law”), it is clear that positive law follows a logic of self-preservation that is hostile to secession. But, if as researchers we agree to take a step back to view the system from an “exterior” point of view (this is the scientific approach, which produces a “discourse about the law”), things appear differently, and the compatibility between federalism and secession appears to depend on political will and the harmonization between two “oughts” (the normative dimension) that can never exist at the same time (the empirical dimension). This involves a manoeuvre or choice that is legally simple and possible, as we have shown.

And what about federalism, or the federal idea? Do the principles of federalism provide, unlike the theory of the classical state, any arguments that support secession? Here, we must clearly distinguish between two ways of conceiving of federalism: as a (decentralized) territorial articulation of power within a single people; or as a special (non-centralized) territorial articulation of power among several different peoples. In other words, we face the well-known distinguo between territorial federalism and pluralist, or pluri-nation, federalism. We must note, in passing, that Carl Schmitt has already pointed to the need for existential homogeneity in a federal-type democracy,70 which he does not hesitate to describe, significantly, as “a federal state without a federal foundation”.71 This type of federalism generally takes a poor view of secession, but this is not true for pluralist or pluri-nation federalism, in which existential pluralism (or ←145 | 146→multiple peoples), which underlies and gives meaning (at least at first) to the federal relationship, inclines the players to accept that the founding freedom of the peoples, without which the federation would not exist, remains a fundamental principle once the federation has been created. A failure to recognize the right of withdrawal of the federated parties could be seen as a derogation from the federative pact, as mentioned by Proudhon, which the founding peoples would possibly not have signed if they had thought, even for an instant, that the federation would one day become a prison.

1 This applies, as we will see, to the authors who define federalism by emphasizing the contractual nature of the relationship or federative logic, while at the same time refusing the principle of secession. For a concrete example, see the following footnote.

2 In a classical work of militant federalist literature, Voyenne, B., Histoire de l’idée fédéraliste, t. III, Paris-Nice, Presses d’Europe, 1981, p. 131–158, the author returns to Calhoun’s famous dilemma: “Either the treaties that led to the state’s birth continue to apply, in which case the state exists only by virtue of the sovereignty of its members; or else, as the federalists claim, the compact is definitive, in which case the state is actually a unitary state” (p. 134). And as a conclusion: “[…] a true federation can only, in our opinion, refuse […] to explicitly recognize a right [of secession]” (p. 153). According to Voyenne, the federative compact is definitive (perpetual) and the member states no longer have the freedom to leave the federation.

3 In keeping with the celebrated comment by Jellinek about the impossibility of legally dissolving a union under public law based on the mere will of its members (“political suicide is not a juridical category”, Allgemeine Staatslehre, 3rd ed. Athenaeum, 1911, p. 768; French translation: L’État moderne et son droit, t. II, Paris, Panthéon-Assas, 2005, p. 538–539), the dominant doctrine is hostile to secession, as shown in the following examples from specialists in the area of federalism: Watts, R. L., New Federations: Experiments in the Commonwealth, Oxford, Clarendon Press, 1966, p. 312; King, P., Federalism and Federation, London, Croom Helm, 1982, p. 112; Bowie, R. R., Friedrich, C., Études sur le fédéralisme, vol. II, Paris, L.G.D.J., 1960, p. 770. It can also be seen as highly significant, in this connection, that one of the most far-ranging and complete studies of recent years (Beaud, O., Théorie de la Fédération, Paris, Presses universitaires de France, 2007) does not develop the question of secession, despite the fact that the author’s theoretical construction, critical of the dominant doctrine and focused on the contractual nature underlying the federal relationship (and also on the freedom of the contracting parties), would have provided a perfect opportunity for an in-depth examination of the question of secession.

4 See, on this point, the criticism and developments of Beaud, O., op. cit., p. 67 ff.

5 See Carl Friedrich, for example: “There can be no sovereign in a federal system, a political order in which autonomy and sovereignty are mutually exclusive” (Friedrich, C., Tendances du fédéralisme en théorie et en pratique, Bruxelles, Institut belge de science politique, 1971, p. 19). It is important to note that this thesis is not merely descriptive (or intended as such). Many other specialists of federalism also believe that one cannot (or must not) consider federalism and sovereignty at the same time, since the two concepts are antinomic. Even Olivier Beaud, the author who, in my view, has produced the most far-ranging and rigorous review of federalism in recent years, shares this opinion: “It is not enough to claim that the theory of sovereignty is not a suitable tool for thinking about federalism and federal norms. In my view, it is necessary to go further by putting forward the idea that, far from being the basic condition for a study of federalism, sovereignty is, on the contrary, the obstacle that must be removed in order to think about federalism” (ibid., p. 58–59, emphasis original; see the development of this question p. 37–97). I, in turn, believe that it is pointless trying to think about federalism while ignoring or sidestepping the notion of sovereignty, because that notion, whether we want it to or not, plays the role of the cat that becomes a woman in the fable by La Fontaine: we can drive an inconvenient fact out through the door, but it will always come back through the window. Given this situation, we might just as well include it in the discussion.

6 My criticism of this approach is found in “L’autorité de la doctrine en droit. L’exemple du fédéralisme”, in Cagiao y Conde, J. (ed.), La notion d’autorité en droit, Paris, Le Manuscrit, 2014, p. 101–134.

7 Olivier Beaud describes the difference between internal and external sovereignty as follows: “Internal sovereignty is a power to command that manifests itself in unilateral acts that reflect a relationship of subordination between the author and the receiver of a norm. In a contract, international sovereignty can only be defined as a ‘power’ to command since it is manifested positively in juridical acts (treaties, customs) that require the consent of the receiver of the norm and negatively in the prohibition of norms imposed by other state powers. The notion of sovereignty is therefore asymmetrical: it is absolute in its internal sphere, and relative in its external sphere, where it encounters its alter ego, the sovereignty of the other state” (Beaud, O., La puissance de l’État, Paris, Presses universitaires de France, 1994, p. 16).

8 Some confusion and misunderstandings doubtless come from the meaning selected or preferred for the concept of “sovereignty”, whether weak (sovereignty as a “competence” in law) or strong (deciding without appeal or adapting the law to its will). This is why we can say that a judge is sovereign in his or her function (the competence to state the law), or that a US state is sovereign in the area of civil law, where sovereignty is understood in its weak meaning. We know this because both the judge and the US state can see their “sovereign” will disavowed by a later sovereign decision (a judge in a court of appeal or final jurisdiction, in the first case, or the Government of the United States or the Supreme Court, in the second). If a sovereign is a party able to impose its own will, then the only holder of sovereignty from which no appeal lies is sovereign in the strong and authentic sense. See Beaud, O., La puissance de l’État, op. cit.; Troper, M., “La souveraineté, inaliénable et imprescriptible”, in Troper, M., Le droit et la nécessité, Paris, Presses universitaires de France, 2011, p. 77–98.

9 Beaud, O., Théorie de la Fédération, op. cit.

10 The impact of nationalism (and state nationalism in particular) on the federal idea and the evolution of federalism is well known today in the specialized literature, based on the classical studies of Hobsbawn, Gellner, Smith, Anderson, Billig and others: Norman, W., Negotiating Nationalism. Nation-building, Federalism, and Secession in the Multinational State, Oxford, Oxford University Press, 2006; Burgess, M., Gagnon, A.-G., Federal Democracies, London, Routledge, 2010; Caminal, M., El federalismo pluralista. Del federalismo nacional al federalismo plurinacional, Barcelona, Paidós, 2002; Requejo, F., Caminal, M., Political Liberalism and Plurinational Democracies, London, Routledge, 2011; Máez, R., La frontera interior. El lugar de la nación en la teoría de la democracia y el federalismo, Murcia, Tres Fronteras, 2008; Kymlicka, W., Norman, W. (eds.), Citizenship in Diverse Societies, Oxford, Oxford University Press, 2000; Gagnon, A.-G., The Case for Multinational Federalism. Beyond the All-encompassing Nation, London, Routledge, 2010; Gagnon, A.-G., Minority Nations in the Age of Uncertainty. New Paths to National Emancipation and Empowerment, Toronto, University of Toronto Press, 2014; Requejo, F., Fédéralisme multinational et pluralisme de valeurs. Le cas espagnol, Brussels, Peter Lang, 2009; Seymour, M., Laforest, G. (eds.), Le fédéralisme multinational. Un modèle viable?, Brussels, Peter Lang, 2011; Burgess, M., In Search of the Federal Spirit. New Theoretical and Empirical Perspectives in Comparative Federalism, Oxford, Oxford University Press, 2012. The topic is relatively unexplored in the field of public or constitutional law, although a few studies by legal specialists have appeared in the last fifteen years, in which the authors incorporate lessons from academic debates on nationalism (mainly by historians and political scientists) and debates between classical liberal philosophers and cultural liberal philosophers starting in the 1990s. For example: Brouillet, E., La négation de la nation. L’identité culturelle québécoise et le fédéralisme canadien, Sillery (Québec), Septentrion, 2005; Parent, Ch., Le concept d’État fédéral multinational. Essai sur l’union des peuples, Brussels, Peter Lang, 2011; Bossacoma i Busquets, P., Justícia i legalitat de la secessió. Una teoria de l’autodeterminació nacional des de Catalunya, Barcelona, Institut d’Estudis Autonòmics, 2015; Ferraiuolo, G., Costituzione Federalismo Secessione. Un itinerario, Napoli, Editoriale Scientifica, 2016; Cagiao y Conde, J., Tres maneras de entender el federalismo: Pi y Margall, Salmerón y Almirall. La teoría de la federación en la España del siglo XIX, Madrid, Biblioteca Nueva, 2014.

11 In the meaning given by Bobbio: a discourse about the law. It is important to distinguish clearly between this discourse about the law and the discourse of the law, which is the object studied by legal science. The expression “legal logic” can also refer to two forms of expression: the first scientific (discourse about the law) and the other legal (discourse of the law). Only the second form of logic, in a strictly Kelsenian sense with which we agree, is actually “legal”, in other words uses prescriptive language. But, as we will see below, both forms of logic can converge on a legal conclusion, each in its own way: the first by descriptively recording the various positive law experiences that have made federalism and secession compatible and incompatible – showing that there is neither compatibility nor incompatibility a priori; the second by highlighting that the law, in the Kelsenian explanation, is subject to interpretation; in other words the will of the implementing authority may or may not want federalism and secession to be compatible, depending on the political context.

12 Carl Schmitt pointed this out in his Constitutional Theory, published in 1928: “The connection of democracy and federal state organization leads to a distinctive, independent type of state organization, to the federal state without an alliance foundation.” (Schmitt, C., Théorie de la Constitution, Paris, Presses universitaires de France, 1993, p. 537).

13 On these two concepts, voir King, P., op. cit.

14 See Burgess, M., op. cit.

15 Ibid.

16 Reference re Secession of Québec, [1998] 2 S.C.R. 217.

17 Kelsen, H., Théorie pure du droit, Paris, L.G.D.J., 1999, p. 335 ff.

18 See Troper, M., Pour une théorie juridique de l’État, Paris, Presses universitaires de France, 1994; Troper, M., Le droit et…, op. cit.

19 I developed this point in a recent study of the opinion of Spanish legal specialists with respect to the Catalan independence referendum and its possible incorporation in the Spanish constitutional order. I explained that most Spanish constitutionalist doctrine has adopted a position on the independence referendum that appears to move away from an approach based on legal positivism (in the sense given by Bobbio) and closer to the specific form of legal positivism found in legal systems. In this way, the dominant doctrine relies on the “discourse of the law” that places it at a distance from the “discourse about the law” used in legal science. Cagiao y Conde, J., “¿Es posible un referéndum de independencia en el actual ordenamiento jurídico español? El derecho explicado en la prensa”, in Cagiao y Conde, J., Ferraiuolo, G. (eds.), El encaje constitucional del derecho a decidir. Un enfoque polémico, Madrid, La Catarata, 2016, p. 142–182.

20 Ibid.

21 Texas vs White, 74 U.S. 7 Wall. 700 700 (1868). According to Cass Sunstein, “no serious scholar or politician now argues that a right to secede exists under American constitutional law” (Sunstein, C., “Constitutionalism and Secession”, The University of Chicago Law Review, vol. 58, n° 2, 1991, p. 633).

22 Dechâtre, L., Le pacte fédératif européen, thesis, Université Paris Panthéon Sorbonne, 2012, p. 42.

23 Beaud, O., Théorie…, op. cit., p. 266.

24 The same criticism is found in Dechâtre, L., op. cit. See the development and nuances introduced by Beaud on the topic of secession from a federation, in this book.

25 See Feldman, J.-Ph., La bataille américaine du fédéralisme. John C. Calhoun et l’annulation (1828–1833), Paris, Presses universitaires de France, 2004; Zoller, E., “Aspects internationaux du droit constitutionnel. Contribution à la théorie de la fédération d’États”, Recueil des Cours de l’Académie de La Haye, t. 294, 2002, p. 41–166; Robel, L., Zoller, E., Les États des Noirs. Fédéralisme et question raciale aux États-Unis, Paris, Presses universitaires de France, 2000.

26 Cited by Sunstein, C., “Constitutionalism and Secession”, The University of Chicago Law Review, vol. 58, n°2, 1991, p. 633, 657.

27 Schmitt, C., op. cit., p. 520.

28 See Beaud, O., Fédéralisme et Fédération en France. Histoire d’un concept impossible?, Strasbourg, Presses universitaires de Strasbourg, 1999, p. 7–82.

29 Proudhon, P.-J., Du principe fédératif, Antony, Tops-Trinquier, 1997, p. 88.

30 Ibid.

31 Ibid., p. 90.

32 “The contract of federation […] is essentially limited. The authority responsible for its execution can never overwhelm the constituent members; that is, the federal powers can never exceed in number and significance those of local or provincial authorities, just as the latter can never outweigh the rights and prerogatives of man and citizen. If it were otherwise, the community would become communistic; the federation would revert to centralized monarchy; the federal authority, instead of being a mere delegate and subordinate function as it should be, will be seen as dominant; instead of being confined to a specific task, it will tend to absorb all activity and all initiative; the confederated states will be reduced to administrative districts, branches, or local offices. […] The same will hold, with even greater force, if for reasons of false economy, as a result of deference, or for any other reason the federated towns, cantons or states charge one among their number with the administration and government of the rest. The republic will become unitary […].” (ibid., p. 87–88).

33 Buchanan, A., Secession. The Morality of Political Divorce from Fort Sumter to Lithuania and Quebec, Boulder, San Francisco, Oxford, Westview Press, 1991.

34 Norman, W., “The Ethics of Secession as the Regulation of Secessionist Politics”, in Moore, M. (ed.), National Self-Determination and Secession, Oxford, Oxford University Press, 1998, p. 34–62; Norman, W., Negotiating Nationalism. Nation-Building, Secession and Federalism in the Multinational State, Oxford, Oxford University Press, 2006; Lehning, P. B. (ed.), Theories of Secession, London, Routledge, 1998; Beran, H., “A Liberal Theory of Secession”, Political Studies, vol. 32, n°2, 1984, p. 21–31; Beran, H., “A democratic theory of political self-determination for a new world order”, in Lehning, P. B. (ed.), op. cit., p. 32–59; Bauböck, R., “Why stay together? A pluralist approach to secession and federation”, in Kymlicka, W., Norman, W. (eds.), Citizenship in Diverse Societies, Oxford, Oxford University Press, 2000; Weinstock, D., “Vers une théorie normative du fédéralisme”, Revue internationale des sciences sociales, n°167, mars 2001; Sorens, J., Secessionism: Identity, Interest, and Strategy, Montreal & Kingston, McGill-Queen’s University Press, 2012; Casesse, A., Self-Determination of Peoples: A Legal Reappraisal, Cambridge, Cambridge University Press, 1995.

35 See Buchanan, A., op. cit.; Bossacoma, P., op. cit.

36 For example: Bastida, X., “El derecho de autodeterminación como derecho moral: una apología de la libertad y del deber político”, in Cagiao y Conde, J., Ferraiuolo, G. (eds.), El encaje…, op. cit., p. 218–268. The French version of this text was published in Cahiers de civilisation espagnole contemporaine: “Le droit à l’autodétermination comme droit moral.”

37 Burgess, M., op. cit.

38 Weinstock, D., op. cit.

39 Norman, W., Negotiating nationalism…, op. cit.

40 Jovanović, M., Constitutionalizing Secession in Federalized States. A Procedural Approach, Portland, Eleven International Publishing, 2007.

41 Weinstock, D., op. cit.

42 Sorens, J., op. cit., p. 110.

43 On this point, see the study by Christophe Parent in this book.

44 50.58 % for “No” and 49.42 % for “Yes”, with a participation rate of 93.5 %.

45 Reference re Secession of Québec, op. cit., par. 2.

46 Ibid., par. 92.

47 Ibid., par. 94 ff.

48 Ibid.

49 An Act to Give Effect to the Requirement for Clarity as Set out in the Opinion of the Supreme Court of Canada in the Quebec Secession Reference, S.C. 2000, c. 26.

50 The Supreme Court emphasizes, for example, the obligation on both parties to negotiate in good faith, which seems to be an invitation to look for negotiated solutions. However, the Canadian parliament has reserved a unilateral right to decide whether a clear majority has voted for independence. A situation could, therefore, arise in which 52 % of Québec’s population votes for independence and at a later date the House of Commons considers that this is not a clear majority for the purposes of secession, which would block the negotiation process and secession itself.

51 An Act Respecting the Exercise of the Fundamental Rights and Prerogatives of the Québec People and the Québec State, CQLR, c. E-20.2.

52 Section 4, for example, contradicts the Clarity Act in terms of the majority needed for a referendum victory: “When the Québec people is consulted by way of a referendum under the Referendum Act (chapter C-64.1), the winning option is the option that obtains a majority of the valid votes cast, namely 50 % of the valid votes cast plus one.”

53 It is important to remember that the EU faced the same questions as far more integrated political contexts concerning the possible withdrawal of a member state. Some of the doctrine suggested that there was no right of withdrawal before it was explicitly recognized in article 50 of the Lisbon Treaty in 2009. This recognition did not prevent some observers from expressing doubts and regrets during the Brexit debate.

54 Constitution of the Republic of Uzbekistan.

55 Act on Greenland Self-Government.

56 On this point, see the explanation of Christophe Parent in this book.

57 “Peace treaties, trade agreements, treaties or agreements relating to international organization, those committing the finances of the State, those modifying provisions which are the preserve of statute law, those relating to the status of persons, and those involving the ceding, exchanging or acquiring of territory, may be ratified or approved only by an Act of Parliament.

They shall not take effect until such ratification or approval has been secured.

No ceding, exchanging or acquiring of territory shall be valid without the consent of the population concerned.”

58 The lack of a written constitution was suggested to explain the flexibility available to the British government when compared to the Spanish government. However, Canada has a written constitution, as do France and Denmark.

59 Spain’s problem in Catalonia since 2012 has given observers an opportunity to view a range of positions: sometimes national unity (art. 2) and sometimes the sovereignty of the Spanish People (art. 1) pose, in the doctrine, an insurmountable legal obstacle not only to the organization of a (consultative) referendum on independence, or independence declared without a referendum, but also to the introduction into the Spanish constitution of concepts or expressions (“nation of nations”, “plurinationality”) proposed by the political parties that support a federal reform of the central state (on the centre-left and left, respectively the PSOE and Podemos).

60 Bobbio, N., Essai de théorie du droit, Paris, Bruylant-L.G.D.J., 1998, p. 24.

61 See Troper, M., op. cit.; Bobbio, N., ibid.

62 Bobbio, N., ibid., p. 24.

63 Ibid.

64 See Kelsen, H., Théorie générale des normes, Paris, Presses universitaires de France, 1996. One of the key questions is whether this intention can be known, a point that divides legal specialists grosso modo into two groups: cognitivists and noncognitivists. For the cognitivists, the intention hidden in a legal statement or norm can be known (by a judge or by a legal professor), while the non-cognitivists defend the opposing thesis: it is impossible to know the intention (meaning) enclosed in a normative statement by the organ that created it. This leads to an equally differing conception of legal interpretation: for the first group, interpretation is a function of knowledge, and the law always expresses the intention of the legislator (the organ that created the norm being interpreted); for the second group, interpretation is an act of will and it is the will of the judge, rather than the legislator, that counts. See Troper, M., Le droit…, op. cit., p. 155 ff.

65 Kelsen, H., op. cit., p. 288.

66 Kelsen, H., General Theory of Norms, Oxford Scholarship Online, 2012, chapter 29. DOI:10.1093/acprof:oso/9780198252177.003.0029.

67 There is no conflict because the absence of a demand for a right of withdrawal from a federation indicates indirectly that the norm to be followed is the norm that says that unity must be.

68 Kelsen, H., op. cit., p. 289 (our translation).

69 Kelsen states that only the absence of a sanction actually represents an infringement of the legal norm. See Kelsen, H., Théorie générale…, op. cit., p. 175–176.

70 Cagiao y Conde, J., “L’intégration fédérale dans l’UE et les leçons de l’histoire: Madison, C. Schmitt et Proudhon”, Revue d’ études proudhoniennes, 2, 2016, p. 71 ff.

71 Schmitt, C., op. cit., p. 537.